Complications from chronic hepatitis C (HCV) and recurrent HCV post-transplant are responsible for significant morbidity and mortality in the United States and Europe. Current antiviral therapies are at best, effective in up to 50% of patients in the pre-transplant setting, and in the post-transplant setting are associated with more limited efficacy and increased toxicity. With this reduced efficacy of antiviral strategies in the post-transplant setting, new approaches are urgently needed. Substantial progress has been made in understanding the pathogenesis of hepatic fibrosis over the last 20 years, which has yielded potential new therapeutic targets. The prospect of antifibrotic therapies is nearing reality in order to reduce progression to cirrhosis, thereby reducing morbidity, mortality and the need for re-transplantation. Current and evolving approaches primarily target the activated hepatic stellate cells, which are the main source of extracellular matrix, along with related fibrogenic cell types. Key issues yet to be clarified include the optimal duration of antifibrotic therapies, endpoints of clinical trials, indications in clinical practice and whether combination therapies might yield synergistic activity.
There has been great progress made over the past 20 years in understanding hepatic fibrosis, the liver's wound healing response and in particular the importance of the hepatic stellate cell (HSC) as the central component in the fibrogenic process. When fibrosis advances to cirrhosis the life threatening complications associated with liver disease occur, including variceal bleeding, ascites formation and hepatorenal syndrome, among others. These complications create a substantial burden on healthcare resources and are expected to increase dramatically in the near future (1). Efforts are underway to target the HSC in an effort to attenuate the fibrotic response. Such approaches may effectively slow fibrosis progression, or could even reverse it. Ideally, treatment with antifibrotic compounds could also lead to cirrhosis regression.
This review will highlight fibrosis progression in patients with chronic hepatitis C (HCV) pre-orthotopic liver transplantation (OLT) and recurrent HCV post-OLT, review the pathogenesis of fibrosis and discuss targeted antifibrotic therapies directed against parts of the HSC activation cascade.
Chronic HCV Infection in the Pre- and Post-Transplant Setting
The development of antifibrotic therapy is especially important for patients with chronic HCV and those with recurrent HCV and fibrosis post-OLT. HCV is the most prevalent blood-borne viral disease in the United States: approximately 4,000,000 people are seropositive for HCV and 270 000 are chronically infected with HCV (2). Seventy four percent of patients with HCV in the United States are infected with Genotype 1, which is least likely to respond (up to 50%) to the only drugs approved by the FDA for treatment of this infection: pegylated interferon with ribavirin (3). End-stage liver disease associated with chronic HCV infection is the leading cause for decompensated liver disease and is the leading indication for OLT in the United States. Within the next 4 years, the projected need for liver transplantation will increase by 528%, mainly due to the HCV epidemic (1). While the need for OLT is increasing, the number of deceased donor organs available has effectively plateaued. Living donor liver transplantation has not significantly impacted on mortality in patients with end-stage liver disease. Thus, there is a dire need for new approaches to treatment; antifibrotic therapies are amongst the most promising.
The development of antifibrotic therapies is additionally important in patients with HCV recurrence post-OLT. Approximately 50% of liver transplants in the United States and in Europe are performed for patients with chronic HCV. Recurrence is virtually universal in those transplanted with active infection. In a majority of patients, disease recurrence is mild or slowly progressive. However, many patients develop progressive injury to their liver allograft with fibrosis progression post-transplant, advancing at a rate that is almost 10-fold higher than fibrosis progression in the pre-transplant setting. Approximately 25–33% of these patients develop cirrhosis within 5 years of undergoing OLT (4–6) contrasted with the pre-OLT setting where cirrhosis requires ∼20–40 years to develop. Thus, progression to cirrhosis in patients with recurrent HCV post-OLT is accelerated compared to the pre-OLT setting (6,7). Progression to cirrhosis post-OLT is near-linear in a significant percentage of patients with a median rate of fibrosis progression of 0.3 Metavir units/year, ranging from 0.6 to 0.8 stage/year (6–8). A subgroup of patients develop fibrosing cholestatic hepatitis, characterized by very aggressive recurrence, and leading to graft failure in 50% of patients within a few months (6).
Once cirrhosis occurs, the rate of decompensation in transplanted patients is accelerated compared to non-transplanted cirrhotic patients. At the first and third years, the rate of decompensation is 42% and 63%, respectively, in transplanted patients, compared to 3% and 18% of non-transplanted cirrhotic patients who decompensate (9). Ultimately, this rapid fibrosis progression leads to reduced graft and patient survival when compared to patients who are transplanted for non-viral causes (11,12). The 5-year survival rates of 60–70% for patients who have been transplanted for HCV-associated liver disease are less in those patients who have undergone transplantation for other causes, in whom survival is 75–80% (10,11). Antiviral therapy for recurrent HCV post-OLT has met with limited efficacy and substantial toxicity (11).
Factors that may be associated with the more rapid progression of fibrosis post-OLT include donor age, immunosuppression, cytomegalovirus infection, race, with non-Caucasians doing worse than Caucasians, HCV RNA levels pre- and post-OLT and genetic background of the recipient (10).
What Is Fibrosis?
Fibrosis is the liver's scarring response to injury that occurs in most chronic inflammatory liver diseases, including hepatitis B (HBV) and HCV, autoimmune or metabolic, for example, hemochromatosis. The ultimate result of chronic injury is the accumulation of extracellular matrix (ECM) and replacement of low density type IV collagen with high density type I collagen within the subendothelial space of Disse (12) (Figure 1). With prolonged chronic injury cirrhosis may develop, which connotes the distortion of normal architecture with nodule formation surrounding regenerative hepatocytes. The major complications of end-stage liver disease occur almost exclusively in patients with cirrhosis, because as more scar matrix is deposited within the subendothelial space of Disse there is disruption of the intercellular milieu, eventually leading to disruption of the cellular homeostasis and culminating in hepatocellular dysfunction. This dysfunction is manifested clinically by complications of portal hypertension (ascites, variceal hemorrhage, portosystemic encephalopathy) and liver synthetic dysfunction. The hope is that if antifibrotic therapy can reconstitute the normal microenvironment of liver, normal function can be restored and clinical manifestations may regress.
Fibrosis requires many years to decades of injury to develop in most patients (13), but there are at least three settings where progression is typically more rapid: infants with congenital hepatic fibrosis may present at birth with advanced fibrosis, patients with recurrent HCV post-OLT, patients co-infected with HCV and HIV. Mechanisms underlying rapid fibrosis development in these cases are unknown, but may reflect profound alterations in matrix degradation and/or significant immune dysfunction, which is increasingly recognized to regulate fibrosis independent of effects on injury.
Fibrosis is generally reversible if the primary injury is removed. In contrast, cirrhosis has historically been viewed as irreversible; however, recent animal and human clinical data in patients with cirrhosis due to HBV (14) and HCV (15) have demonstrated that cirrhosis may be reversible when the underlying chronic injury is removed. At some point, it is clear that cirrhosis becomes truly irreversible, as manifested clinically by refractory ascites, encephalopathy, hyponatremia and renal dysfunction, but the feature(s) which distinguish irreversible fibrosis are yet to be identified. Recent animal studies have suggested that cirrhosis becomes irreversible when collagen cross-linking by tissue transglutaminase occurs, which probably leads to an insoluble hepatic matrix characterized by scar that is relatively resistant to degradation (16). Additional factors which may influence the reversibility of cirrhosis might include (a) total collagen content, which may coalesce into broad fibrotic bands that are inaccessible to collagen degrading enzymes; (b) duration of cirrhosis, which could reflect a longer period of collagen cross-linking, rendering it less sensitive to degradation by enzymes over time and (c) reduced expression of matrix-degrading enzymes, or increased expression of proteins which inhibit matrix-degrading enzyme function.
Fibrogenesis and the Role of the Hepatic Stellate Cell (HSC)
Central to fibrogenesis is the activation of hepatic stellate cells (HSC). HSC reside in the space of Disse between hepatocytes and sinusoidal endothelial cells. In normal liver, HSC are quiescent and store vitamin A. In injured liver, HSC become ‘activated’, in which quiescent, vitamin A-rich cells transit into myofibroblasts capable of proliferation, fibrogenesis, contractility and matrix degradation, among other functions (12,13) (Figure 2). Activated HSC are also a key source of mediators, matrix molecules, proteases and their inhibitors that together lead to the formation of the liver scar. Each step of the ‘activation’ cascade may be targeted by an antifibrotic compound, and is detailed below.
PDGF-BB (platelet-derived growth factor beta) is the most potent proliferative stimulus toward HSC. Both PDGF-BB and its receptor are upregulated following liver injury (17). Other mitogenic factors and their cognate tyrosine kinase receptors which increase HSC numbers during liver injury include thrombin, insulin-like growth factor, endothelin-1, fibroblast growth factor and vascular endothelial growth factor (VEGF) (18,19). Thus, targeting activated stellate cell proliferation with an inhibitor of tyrosine kinases may attenuate fibrosis.
HSC, Kupffer cells and platelets secrete transforming growth factor beta-1 (TGFβ-1), the most potent fibrogenic factor for HSC (20,21). Both quiescent and activated HSC express TGFβ-1 receptors (22).
Activated HSC are contractile, primarily in response to endothelin-1 (ET-1) (23). Net contractile activity of HSC is determined by the balance between ET-1 and its antagonist nitric oxide (NO) (24). In advanced chronic liver disease, there is increased ET-1 and decreased NO activity. Furthermore, because of their location in the subendothelial space of Disse, activated HSC may constrict individual sinusoids, thereby increasing portal resistance.
In normal liver, the rate of ECM production equals that of its degradation, resulting in no net accumulation of matrix. Fibrogenesis occurs when there is an imbalance between ECM degradation and production. Proteinases, including the matrix metalloproteinases (MMP), are responsible for matrix breakdown and include five categories: interstitial collagenases (MMP-1, -8, -13), gelatinases (MMP-2, -9), stromelysins (MMP-3, -7, -10, -11), membrane type (MMP-14, -15, -16, -17, -24, -25) and a metalloelastase (MMP-12) (25). Regulation of metalloproteinases is complex. Inactive MMPs may be activated through proteolytic cleavage and inhibited by binding to specific inhibitors known as tissue inhibitors of metalloproteinases (TIMPs). Downregulation of MMPs by a reduction in gene expression or an increase in their tissue inhibitors favors the accumulation of ECM. Progressive fibrosis is associated with increases in TIMP-1 and TIMP-2, which leads to net decrease in MMP (protease) activity, and therefore more unopposed matrix accumulation (26,27). HSC are the major source of TIMPs.
Cytokine release and chemotaxis
HSC release cytokines including TGFβ-1, PDGF-BB, fibroblast growth factor, hepatocyte growth factor (HGF), platelet activating factor, ET-1 and neutrophil and monocyte chemoattractants which can amplify inflammation in liver injury. HSC also release anti-inflammatory cytokines like interleukin-10. HSC may also migrate directly to areas of injury in response to PDGF (the most potent chemotactic factor), endothelin and monocyte chemotactic protein 1 (28,29).
Loss of vitamin A (retinoids)
Activated HSC lose their vitamin A droplets, but it is not known if vitamin A loss is required for HSC activation.
Resolution of liver fibrosis and the fate of activated HSC
It is unclear what happens to HSC when fibrosis and cirrhosis resolve. There is a decrease in the number of activated HSC as fibrosis resolves, but it is uncertain if HSC return to a quiescent state or undergo apoptosis (30). Cell culture has demonstrated that the HSC can revert to a quiescent cell type, but it is unknown if this occurs in vivo (28). Inducing apoptosis of activated stellate cells is possibly another targeted antifibrotic.
Potential Antifibrotic Therapies
Properties of an antifibrotic compound
Since hepatic fibrosis pathways are largely similar in all liver diseases regardless of their etiology, the development of antifibrotics should benefit all patients with fibrosing liver injury. Because of the liver's great ability to regenerate, liver fibrosis is an attractive target for antifibrotic therapy compared to fibrosing diseases in other organs like the lung and kidneys, which tend to have less regenerative capacity. Furthermore, because of first-pass metabolism through the liver, lower doses of orally available compounds may be necessary to achieve a therapeutic response, thereby minimizing systemic distribution of the agent and non-hepatic side effects.
Desirable properties of an antifibrotic compound include the following features: (a) HSC-specific and non-toxic to neighboring hepatic cells; (b) no effect on the metabolism of immunosuppressive agents, to avoid increasing the toxicity of immunosuppressive agents on the kidneys or nervous system by increasing their relative activity or accelerating the chance of rejection if relative drug levels are reduced;(c) orally available and easily delivered; (d) non-toxic to other organs, such as the kidneys and (e) safe when used over long periods of time.
Obstacles to antifibrotic drug development
There are still some obstacles to the development of antifibrotic therapies. Since fibrogenesis is a slow process that requires years or decades to develop, assessment of changes over a short interval is problematic. Furthermore, prospective human clinical trials of potential antifibrotic agents may be costly because of long duration. What is the optimal duration of therapy? Typically, pharmaceutical companies appear willing to sponsor 1-year ‘proof of concept’ trials. However, is 1 year enough to demonstrate an antifibrotic effect of a putative compound, given that fibrosis takes years to develop? Furthermore, an anti-fibrotic trial would probably require a pre- and post-treatment biopsy to assess the antifibrotic efficacy, since current non-invasive tests for liver fibrosis cannot replace biopsy.
Nonetheless, there are limitations to a liver biopsy, even though this technique is considered the gold standard for assessing hepatic fibrosis. The procedure is invasive and adverse events including pain, hemorrhage necessitating intervention, rupture of adjacent organs (0.3–0.5%) and death (0.01–1%) occur (31–33). As a result, many patients are reluctant to undergo this procedure. Historically, about 33% of patients in clinical trials have refused post-treatment liver biopsies (34,35). The ability to perform repeated measures of fibrosis to assess treatment efficacy is vital to the success of any antifibrotic trial.
Sampling error is the major limitation of liver biopsy (36–40). In a study of 124 patients with HCV who underwent a laparoscopic biopsy, 33% had a difference of at least one fibrosis stage between biopsies from the right and left hepatic lobes (41). Clearly, a measure of fibrosis that more accurately reflects fibrogenic events occurring within the liver as a whole is needed. In addition to sampling error, liver biopsy is prone to subjectivity in pathological interpretation. Furthermore, despite fibrogenesis being an active process characterized by deposition and degradation of matrix, the liver biopsy reveals only static information about fibrosis in a small fraction of the organ. In contrast, there may be early changes in fibrotic marker gene expression that precede changes in fibrosis as assessed using scoring systems that include Metavir or Knodell scales, and could be used as the basis for novel diagnostic approaches.
Which group of patients would be suitable for antifibrotic therapies? Currently, the main focus is on patients with chronic HCV who have failed antiviral therapy, since they represent a very large cohort with a well characterized natural history. Other target population include those with recurrent HCV and fibrosis post-OLT.
Targeted antifibrotic therapies
There is a tremendous interest in the field of drug development and testing for hepatic fibrosis. To date however, no drugs are approved as antifibrotic therapy. Removing the primary injury is the best antifibrotic therapy. For example, in patients with chronic HCV, treatment with an antiviral in an effort to eradicate the virus is the optimal approach to reversing fibrosis. If the primary disease cannot be cured, then an antifibrotic compound can potentially target any part of the HSC activation cascade described above: (a) downregulating HSC activation; (b) neutralizing proliferative, fibrogenic and contractile responses of HSC; (c) increasing the degradation of scar matrix and (d) stimulating apoptosis of HSC.
Downregulating HSC activation: Attenuating HSC activation is an attractive antifibrotic target. Both gamma interferon (γ-IFN) and HGF have demonstrated inhibitory effects on HSC activation in animal models of liver fibrosis (42,43). A recent double blind controlled clinical trial assessing the antifibrotic efficacy of γ-IFN showed no benefit, however.
Peroxisome proliferator activated nuclear receptor gamma (PPARγ) is expressed in HSC, and synthetic PPARγ ligands, or thiazolidinediones, downregulate HSC activation (44–46). There are trials underway assessing the effect of these ligands in patients with non-alcoholic fatty liver disease, and fibrosis is a major endpoint of such trials.
Oxidant stress stimulates HSC activation, and thus reducing oxidant stress is a possible antifibrotic target. In studies of experimental or human fibrogenesis, antioxidants, including vitamin E, suppress fibrogenesis (47). Herbal products like Silymarin, the active ingredient in milk thistle, as well as the traditional Chinese/Japanese plant extract Sho-saiko-to have been noted to have free radical-scavenging ability, Sho-saiko-to has antifibrotic properties in vitro and in rat models of porcine-serum induced fibrosis (48).
Neutralizing proliferative, fibrogenic and contractile responses of HSC: Neutralizing the proliferative, fibrogenic and contractile responses of HSC is another targeted approach of a potential antifibrotic compound. Cytokine receptor antagonists may be used as potential antifibrotic agents. In activated HSC, the most potent proliferative factor is PDGF-BB, binds to its tyrosine kinase receptor, ββ-PDGF, which is upregulated during activation. Glivec, a targeted tyrosine kinase receptor antagonist, already in use against chronic myelogenous leukemia and gastrointestinal stromal tumors (49), has attenuated the fibrotic response in animal fibrotic models (unpublished data).
The most potent fibrogenic factor for HSC is TGFβ-1 and targeting TGFβ-1 might greatly down-regulate matrix production. The latent form of TGFβ-1 must be cleaved by proteases in order to become active. Different strategies have been used to block activation of TGFβ and to prevent its binding to specific receptors, which demonstrate decreased fibrosis in vivo. Camostate mesilate, a protease inhibitor, inhibits the release of active TGFβ-1 and attenuates liver fibrosis in rats (50). Attempts to prevent the binding of TGFβ to its receptors have successfully attenuated fibrosis in rats. Studies including use of a dominant-negative type II TGFβ receptor, the expression of truncated type II receptor and the construction of a soluble type II receptor have demonstrated antifibrotic efficacy (51,52). There is concern, however, that with long-term TGFβ antagonism in humans, the modulation of inflammation and the immune response and the loss of TGFβ-mediated growth suppression could stimulate hepatocellular growth, thereby promoting cancer.
Neutralizing the effects of the renin-angiotensin system (RAS) is an additional antifibrotic approach. The RAS plays a role in disease states where chronic injury, inflammation and tissue remodeling occur (53). Its main effector, angiotensin II stimulates collagen deposition (54) mainly through the induction of TGFβ-1 (55). ATII probably plays a role in liver fibrosis. The systemic and hepatic and HSC RAS are active in patients with chronic liver disease (56–58). The mechanism(s) by which ATII is profibrogenic are not fully known in liver, but this mediator binds ATI receptors in HSC and induces contraction and proliferation (58). ATI receptor stimulation may also increase oxidant stress through increased NADPH oxidase. In liver injury, increasing circulating ATII levels accelerates inflammation and fibrosis (59). Moreover, inhibiting the RAS attenuates fibrosis development in rats (60); RAS antagonism with losartan, an angiotensin converting enzyme inhibitor, attenuates fibrosis in rats (61,62). Losartan has reversed fibrosis in patients with renal fibrosis, whose pathogenesis is very similar to patients with liver fibrosis (63). Because patients post-OLT often develop systemic hypertension, use of losartan is particularly attractive compared to other antihypertensives, since it may have both hypotensive and antifibrotic activities.
Hepatocyte growth factor (HGF) can attenuate fibrosis in animal models (43). HGF modulates HSC proliferation, collagen formation and TGFβ expression. Similar to TGFβ antagonists, there is concern about the use of HGF over the long term as it may induce or promote the growth of hepatocellular carcinoma (HCC), particularly in cirrhotics, since they already have an increased HCC risk.
Endothelin-A receptor (ET-AR) antagonists are another potential targeted therapy. ET-AR mediates HSC contraction, proliferation and may also stimulate collagen synthesis. In murine models of fibrosis, ET-AR antagonists have both antifibrotic efficacy and can decrease portal pressure, either by reducing endothelin or by increasing nitric oxide (64).
Promoting matrix degradation: Targeting matrix degradation is another antifibrotic target, given the need to resorb matrix in patients with established fibrosis or cirrhosis. TGFβ antagonists stimulate matrix degradation by downregulating TIMPs and increasing activity of interstitial collagenase. Expressing metalloproteinases mRNA via gene therapy in animal fibrotic models has confirmed, in principle, that matrix can be resorbed (65).
Promoting apoptosis: Targeting HSC specifically in the liver for apoptosis is another antifibrotic target. Use of gliotoxin, a fungal toxin that induces apoptosis specifically in HSC, has attenuated the fibrotic response in a rat model of CCl4 induced fibrosis (66).
Significant progress has been made in understanding the role of the HSC in fibrogenesis, which has led to many potential therapies that may improve fibrosis in the pre- and post-OLT setting. Continued refinement in clinical trial design, including optimal choice of subjects, duration of therapy and trial endpoints is anticipated in the immediate future. As the list of agents grows and the determinants of success are more clearly defined, inclusion of antifibrotic therapies is likely to emerge as an important option in patients with fibrosing liver disease, particularly in patients at high risk for progression such as those with HCV post-transplantation.
This study is supported by AASLD Sheila Sherlock Translational Research Award; Wasserman Scholar Award.